Atrial fibrillation (AF) is a complex arrhythmia that afflicts more than 2.3 million Americans. Most prevalent in the elderly, AF is associated with a 2-fold higher risk for mortality. AF is the most potent risk factor for stroke in the elderly and the most common cause of cardioembolic stroke. Despite intensive study, AF treatment options remain suboptimal. Recent data support the hereditability of AF and suggest a complex genetic basis for common AF. Identification of genes that increase risk for AF may have important clinical applicability. The overall goal of our research program is to perform an unbiased whole genome association study to identify genetic variants associated with AF and to gain insight into the pathogenesis of AF. This will promote the identification of new targets for therapeutic intervention and/or diagnostic tests to predict and potentially prevent AF. DNA will be derived from AF patient and control samples collected in DNA/blood/tissue banks at the Cleveland Clinic and collaborating institutions. Our initial case cohort will be comprised of subjects with Lone AF (AF without coronary, valvular, or other structural heart disease), a tightly defined and much more homogeneous phenotype than previously studied mixed etiology AF populations, and likely to have the most direct genetic associations. Using microarray technology, 550,000 specific single nucleotide polymorphisms (SNPs) will be assessed in each of 600 lone AF subjects, 300 matched heart disease-free control subjects, and 1800 publicly accessible genotyped population controls. We will perform multivariate logistic regression to identify the top SNPs associated with AF for subsequent validation. These SNPs will be validated in a 1,536 SNP custom genotyping assay performed with an independent population of 600 subjects with early onset AF (meeting criteria for lone AF) and 600 disease-free population controls. Because most patients with AF have some form of structural heart disease, we will begin to extend the 1,536 SNP analysis to other AF subjects with structural heart disease, specifically well-defined coronary artery disease. We will analyze groups separately, as well as jointly, pooling data from the 550K and 1536 SNP assays from all studies, and perform multiple regression and novel testing strategies to minimize population stratification in order to identify SNPs significantly associated with lone AF and AF with heart disease. We will resequence the genes in the vicinity of the top 4 highly AF-associated SNPs in 48 patients to identify new variations, gain insight into AF-associated haplotypes, and potentially identify causative variations that could cause changes in protein function or expression levels. We will also evaluate the prevalence of copy number variations in lone AF, as well as somatic mutations and copy number variations in atrial tissue from a subset of patients with AF. Future benefits of these studies include increased understanding of the mechanisms of AF pathogenesis and the potential ability to develop personalized, gene-specific diagnostic tests and drugs that will aid in predicting, preventing and treating AF and its associated risks for stroke. Project Narrative: Atrial fibrillation (AF) afflicts more than 2.3 million Americans and is associated with a 2-fold higher risk for mortality and a 5-7 fold increased risk of stroke, making AF one of the most potent risk factors for stroke in the elderly and the most common cause of cardioembolic stroke. Despite intensive study for the past decade, AF treatment options remain suboptimal. Future benefits of the proposed studies to identify genetic variants that increase risk for AF include increased understanding of the mechanisms of AF pathogenesis and the potential ability to develop personalized, gene-specific diagnostic tests and drugs that will aid in predicting, preventing and treating AF and its associated risk for stroke.